DESCRIPTION: (Applicant's Description) Although it has been shown that the global permeability of tumor vessels to blood-borne therapeutic agents is generally high, it is difficult to deliver anti-tumor drugs homogeneously throughout the tumor. This may be due to the slow process of diffusion that governs tumor transvascular transport, combined with the low permeability in localized regions of the tumor vasculature. The presence of these regions in which the endothelial junctions display normal barrier function precludes the passage of macromolecules or particles-the survival of only a few cancer cells in these regions can result in relapse. Problems in drug delivery are also confounded by the organ-specific nature of transvascular permeability. Generalization of therapeutic protocols may not be possible because the transport properties of the vasculature vary depending on the organ in which the tumor resides. Our hypothesis is that these considerations are at least partially responsible for the disappointing results of recent clinical trials of highly promising bio-engineered macromolecules (such as gene targeting vectors and immunoliposomes) and monoclonal antibodies. Another aspect of tumor vasculature that is relevant to drug delivery is the presence of tumor cells interspersed between the endothelial cells of the vessel wall. Our hypothesis is that these "mosaic vessels" may contribute to the observed heterogeneity in drug delivery, and may be potential targets for anti-vascular therapy. The goal of this project is to identify the determinants of paracellular transport of macromolecules and particles across the vascular wall in solid tumors. In vitro and in vivo methods will elucidate the role of cytokines and adhesion molecules responsible for the hyperpermeability of some tumor vessels and the low permeability of others. The site-specific nature of permeability will be studied using endothelial cells from various sites in vitro, and the contribution of mosaic vessels to transvascular transport will be quantified using in vivo tumor models (bearing a GFP gene construct). When completed, the proposed work will provide a fundamental understanding of the mechanisms of transvascular drug delivery that will help in the development of more effective therapies.